wingsweihua / colight

CoLight: Learning Network-level Cooperation for Traffic Signal Control
http://personal.psu.edu/hzw77/publications.html
160 stars 58 forks source link

Reproducing colight's results #33

Open gsavarela opened 3 years ago

gsavarela commented 3 years ago

Hi,

I would like to reproduce the results you have published in your article. I cloned your repo and ran all default configurations: python -O runexp.py

I guess the model it runs by default has the following configurations, (please correct me if I am wrong):

Scenario: Grid 6x6 bidirectional Model Colight Rounds: 100 Count: 3600

Afterwards, I moved the records/ folder to records/initial/ and ran:

python summary_multi_anon.py

I was expecting to find 170.11 travel time but got ave. duration 182.72. Is the reported travel time on the article the same thing as ave. duration? The difference is relatively small, but I just would like to confirm that I am doing the right thing. Am I?

LJW0306 commented 1 year ago

@gsavarela 您好 我最近也在看这一块的代码。可以和您交流一下吗?

wangkang123789 commented 1 year ago

@LJW0306 您好,我可以和您交流一下吗?

LJW0306 commented 1 year ago

可以啊

------------------ 原始邮件 ------------------ 发件人: "wingsweihua/colight" @.>; 发送时间: 2023年11月1日(星期三) 上午9:45 @.>; @.**@.>; 主题: Re: [wingsweihua/colight] Reproducing colight's results (Issue #33)

@LJW0306 您好,我可以和您交流一下吗?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>